BoC–BoE Sovereign Default Database: What’s new in 2021? by David Beers1, 1 Elliot Jones2, Zacharie Quiviger and John Fraser Walsh3 The views expressed this report are solely those of the authors. No responsibility for them should be attributed to the Bank of Canada or the Bank of England. 1 [email protected], Center for Financial Stability, New York, NY 10036, USA 2 [email protected], Financial Markets Department, Reserve Bank of New Zealand, Auckland 1010, New Zealand 3 [email protected], [email protected] Financial and Enterprise Risk Department Bank of Canada, Ottawa, Ontario, Canada K1A 0G9 1 Acknowledgements We are grateful to Mark Joy, James McCormack, Carol Ann Northcott, Alexandre Ruest, Jean- François Tremblay and Tim Willems for their helpful comments and suggestions. We thank Banu Cartmell, Marie Cavanaugh, John Chambers, James Chapman, Stuart Culverhouse, Patrisha de Leon-Manlagnit, Archil Imnaishvili, Marc Joffe, Grahame Johnson, Jamshid Mavalwalla, Philippe Muller, Papa N’Diaye, Jean-Sébastien Nadeau, Carolyn A. Wilkins and Peter Youngman for their contributions to earlier research papers on the database, Christian Suter for sharing previously unpublished data he compiled with Volker Bornschier and Ulrich Pfister in 1986, Carole Hubbard for her excellent editorial assistance, and Michael Dalziel and Natalie Brule, and Sandra Newton, Sally Srinivasan and Rebecca Mari, respectively, for their help designing the Bank of Canada and Bank of England web pages and Miranda Wang for her efforts updating the database. Any remaining errors are the sole responsibility of the authors. 2 Introduction Since 2014, the Bank of Canada (BoC) has maintained a comprehensive database of sovereign defaults to systematically measure and aggregate the nominal value of the different types of sovereign government debt in default. The database draws on previously published datasets compiled by various public and private sector sources. It combines elements of these, together with new information, to develop comprehensive estimates of stocks of government obligations in default. These include bonds and other marketable securities as well as bank loans and official loans, valued in US dollars, for the years 1960 to 2020 on both a country-by-country and a global basis. The database is posted on the BoC’s website and is updated annually in partnership with the Bank of England (BoE). This update of the BoC–BoE database—and updates to come—will be useful to researchers analyzing the economic and financial effects of individual sovereign defaults and, importantly, the impacts on global financial stability of episodes involving multiple sovereign defaults. In this paper, we first highlight new developments in sovereign debt defaults in 2020. We provide high-level details on the 48 percent increase in aggregate sovereign debt in default, which was driven chiefly by defaults on foreign currency bonds by Venezuela, Argentina and Puerto Rico. We also describe functionality improvements to the database. We then update some key insights the database provided regarding the number, size and types of defaults. We give a historical overview of debt defaults, including the increase in problematic debt since 2018, the persistence of defaults in highly indebted poor countries, and the shift in bilateral lending toward non–Paris Club lenders.4 Finally, we provide an update on our continued efforts to include reliable estimates of domestic fiscal arrears in the database. What’s new in 2021? In this year’s update, our findings estimate the total value of sovereign debt in default at US$443.2 billion in 2020 (0.5 percent of world public debt), up US$143.6 billion (48 percent) from the revised total of US$299.6 billion in 2019. This outpaces the 13 percent increase in gross world public debt. The data by major creditor categories show that the increase was driven mainly by foreign currency bonds in default, which rose by US$121.2 billion. This reflected: . new defaults by Argentina, Belize, Ecuador and Suriname . a first-time default on foreign currency bonds by Lebanon . a greater amount of interest arrears from ongoing bond defaults by Venezuela and Puerto Rico5 Local currency debt in default increased by US$0.9 billion. This was due mainly to Iraq’s restructuring of its short-term debt into debt with longer-term maturities and lower coupon rates. Defaults affecting official creditors, notably the Paris Club, China and other bilateral lenders, also increased.6 Their loans accounted for US$19.4 billion of the global increase, with the Debt Service 4 The permanent members of the Paris Club are Australia, Austria, Belg ium, Brazil, Canada, Denmark, Finland, France, Germany, Ireland, Israel, Italy, Japan, Korea, the Netherlands, Norway, Russia, Spain, Sweden, Switzerland, the United Kingdom and the United States. For more information, see the Paris Club website. 5 Smaller defaults involve not-yet-completed exchanges of old Argentine defaulted bonds and non-performing bonds issued by Nauru and Zimbabwe. 6 These creditor categories exclude the International Monetary Fund, the International Bank for Reconstruction and Development and the International Development Association. 3 Suspension Initiative (DSSI) playing a relatively modest role.7 The values of defaulted debt in other creditor categories registered smaller changes. Regarding our treatment of countries that participated in the DSSI, in the database we include 2020 debt service deferrals by bilateral official creditors for sovereigns that the International Monetary Fund (IMF) or the World Bank considers to be already in or at high risk of debt distress. The suspension period for debt service payments was originally set to run from May through December 2020 but was later extended through December 2021. Participation in the program by debtor sovereigns is voluntary, and the debt relief provided is intended to be net-present-value neutral. However, our reason for including in the database debt service payments suspended under the DSSI is that without the DSSI, many of these vulnerable sovereigns, some already with other debt arrears, would likely have sought debt relief. Total DSSI debt service deferrals included in the database amount to US$4.7 billion, about 1 percent of the total stock of debt in default we identified globally. Some of the changes in this update are: . additional data for defaults on China’s official loans since 2000 . updated annual data (where available) for each country’s total central government debt . minor revisions of country and aggregate default data for 1960–2019 . data, by country and globally, on domestic arrears since 2005, with the most comprehensive data in the years 2018–20 (see the discussion of domestic arrears) We also updated two tabs at the bottom of the main database spreadsheet: DATA provides a downloadable format for the global and country default data, and DEBTOTAL provides country data on total government debt stocks. The previously included DOMARS tab is not included in this year’s update because the domestic arrears data are now incorporated under the first tab. Finally, all data are now also downloadable in the CSV, JSON and XML formats. Key insights from the database Since 1960, 146 governments—two thirds of the current universe of 215 sovereigns—have defaulted on their obligations.8 Defaults had the biggest global impact in the 1980s, reaching US$450 billion, or 6.1 percent of world public debt, by 1990. The scale of defaults has fallen substantially since then. Over the past decade, between 0.3 and 0.9 percent of world public debt has been in default. In 2020, the amount was estimated at 0.5 percent. Total sovereign debt in default increased by 48 percent in 2020, considerably outpacing the 13 percent increase in gross world public debt. 7 In response to the global COVID-19 economic and financial shock, in 2020 the G20—in concert with the International Monetary Fund and the World Bank—launched the DSSI, offering temporary relief on debt service payments owed to bilateral official creditors by 73 low-income countries. The G20 also asked private creditors to participate on comparable terms but was rebuffed. For more information about the DSSI, see World Bank Group (2021). 8 While 154 sovereigns are listed in the database, 6 (Bahamas, Namibia, Palau, Sint Maarten, Tuvalu, and West Bank and Gaza) include only domestic arrears, which we consider to be effective defaults, albeit not on conventional sovereign obligations. For more information on domestic arrears, see page 11 of this paper. 4 As in recent years, the distribution of defaults in 2020 is highly skewed in terms of value. Just three sovereigns—Venezuela, Argentina and Puerto Rico—accounted for 38 percent of the US-dollar value of debt in default globally. The top 10 sovereigns in default accounted for 87 percent. Our database includes the debt owed to official creditors, and this provides insight into sovereign default “clusters,” which we define as spikes in the number of defaults followed by sharp declines. The data show that while the US-dollar amounts can be low in absolute terms, defaults to official creditors often take longer to resolve than defaults involving private creditors. A high number of low-income sovereigns often remain in default to official creditors for long periods. As a percentage of total government debt, only 50 sovereigns—6 percent of observations— defaulted on shares ranging between 50 and 100 percent of their total outstanding government debt. In contrast, about 72 percent of observations are equal to or below 10 percent of government debt. These data confirm sovereigns’ tendency to “default selectively” and show how shares of sovereign debt in default are skewed toward lower values. Despite an uptick in 2020, over the past decade defaults involving the Paris Club group of official creditors have declined in importance. Defaults involving other bilateral official creditors, however, most notably China, are growing.